Simulation And Optimal Control Of Chiller Station System Of Airport Terminal Building

Since 90's, with the rapid development of local regional economy, progress of aerospace technology, and strong growth of aviation needs, the large-scale modernized airport terminal building is becoming an indispensable component of the urban economic connection. And the improvement of living standards makes people not only need a better indoor environment quality of large public transportation buildings when they get away, but also pay more attention to the issues of energy saving and environment protection, which brings the air-conditioning system design more challenges and issues. Among them, in addition to designing the various components of the HVAC (Heating, Ventilation and Air Conditioning) system, the other critical issue is about the control strategy of HVAC system– how to optimize the controls of the system would become one of the most important research subjects of large-scale HVAC system.For the study of energy consumption and HVAC system operation, the simulation has been considered to be a powerful and easy-to-use research tool. It is usually used to predict building energy consumption, diagnose HVAC system fault and optimize HVAC's control strategy. For the large-scale HVAC system in particular, any proposed optimal control strategy should be on the basis on an accurate simulation model, which is approximate to the actual system. Then, it is generally considered to be the cornerstone of the research on HVAC's energy consumption and optimal control strategy. Therefore, this paper has set up a detailed simulation model, corrected and calibrated by comparing the data collected on the chiller station, to ensure the accuracy of the simulation results.In this paper, a large-scale terminal building and its chiller station system located in south of China is selected as the research object. Through the field investigation and measurement, a large number of physical parameters and specific characteristics about the terminal building structure, the air-side of HVAC system and the chiller station water system are obtained. Through the collation and analysis of these data, the cooling load simulation model of the terminal building and the air-side of HVAC system are set up based on EnergyPlus platform, in order to input the data of the real-time cooling load to the simulation of the chiller station water system. And the on-line simulation of the water system (including the chillers, chilled water circulation and cooling water circulation) is set up based on the TRNSYS platform, in order to analyze and validate the optimal control strategies respectively based on chillers'supply water temperature set-point and load distribution ratio.Chillers'supply water temperature is considered to be an important control variable influencing the energy consumption of the pumps and chillers. Therefore, how to optimize the control variable is the key issue for the HVAC system to save energy. Firstly, this paper presents an optimal control strategy based on the best yearly fixed supply water temperature set-point which is validated and analyzed by the established system simulation. However, when the cooling load is gradually changing, the fixed supply water temperature set-point could not minimize energy consumption. To minimize the energy consumption, this paper presents an on-line optimal strategy of chiller's supply water temperature set-point reset. The simulation test results show that compared with the conventional control strategy (the supply water temperature set-point is 7℃), both the control strategy based on yearly the best fixed supply water temperature set-point and the on-line control strategy of the supply water temperature set-point reset have the significant energy savings, and the on-line control strategy is better for energy savings.In addition, chillers'PLR (Part Load Ratio) is another key control variable influencing the energy consumption of the chillers. For the multi-chiller system, especially for the chillers with the different capacity, they are usually operating under part load condition at most time. Therefore, optimization of the load distribution of each chiller is an effective way to improve the overall COP (Coefficient of Performance) of the multi-chiller system. This paper utilizes the regression method to obtain the relation between each chiller's COP and PLR, and establishes the energy consumption equation of each chiller's PLR. Then, an index–cooling load frequency (0~100%) is introduced to enumerate and analyze the yearly total runtime of the HVAC system. Finally, the Lagrange and GA (Genetic Algorithm) methods are employed respectively to solve the energy consumption equation, to obtain the best PLR and energy consumption of each chiller at the different interval of the cooling load ratio (0~100%). The simulation test results show that compared with the conventional sequence control strategy, optimal control strategies based on GA and Lagrange methods have the significant energy savings. In comparison, the GA is better than the Lagrange in terms of energy saving, because it has the advantage of global optimization without local minimum problem.